Multimodal visualization based on mutual information
Roger Bramon Feixas · LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Científicas) · 2013
Medical imaging and 3D volumetric visualization are a standard routine for patient diagnosis and treatment. Multimodal visualization plays an important role in medical diagnosis and treatment, because it makes findings possible that might be missed when each modality is considered separately. Despite the advantages of multimodal visualization, its application in real medical context is still lacking due to the complexity of the technique. To overcome these limitations, we will focus this thesis on the automation of the fusion step and the main involved processes: the information fusion and the transfer function design. On the other hand, we will study the quality evaluation of a rendering where we will propose an objective quality metric, which can be used to assess and optimize different visualization processes. We will utilize information theory tools in order to tackle all described problems